Development of Models for Effective Asphalt Mix Designs Including High Percentages of RAP and RAS

Abstract

The increasing use of recycled materials has led to balanced mix design (BMD), which combines volumetrics design with performance tests. Achieving air void and voids in mineral aggregate (VMA) to meet the volumetric stage of specifications is currently done through a trial-and-error process that is labor-extensive and time- and resource-consuming. Previous researchers have studied different methods to facilitate achieving optimal volumetrics, with limited success. In addition, past studies did not consider the effect of recycled asphalt pavement (RAP) or recycled asphalt shingles (RAS), especially high RAP/RAS designs. This study evaluated 70 different mix designs that vary in terms of material sources, nominal maximum aggregate sizes (NMAS), design numbers of gyrations, binder properties, and RAP/RAS materials. The significant parameters affecting the mix VMA were identified using stepwise statistical analysis. It was found that the inclusion of RAP/RAS increases VMA, whereas the asphalt film thickness had a parabolic correlation with the VMA. Regression models of VMA and air void were developed based on these parameters and verified with the K-fold cross-validation technique, showing a fitting coefficient of 82% or higher. The verified models show the potential to assist mix designers to effectively design asphalt mixtures, especially those with RAP and RAS.